KR101912446B1 - Method and apparatus for detecting intruder - Google Patents

Abstract

The present invention provides a method and a device for detecting intrusion, which is resistance to environmental noise and can obtain a valid signal corresponding to intrusion through a signal process of raw data without using expensive hardware. A radar sensor collects a digital I signal and a Q signal from a received RF signal, calculates each of absolute values of the digital I signal and the Q signal and convolutes the absolute values of the I signal and the Q signal. Next, the radar sensor calculates an average from the convoluted signal during a set time, sets a first threshold value for detecting the valid signal by applying an index weight filter function to the average, and detects the valid signal for detecting intrusion through a comparison between the average and the first threshold value.

Description

≪ Desc / Clms Page number 1 > METHOD AND APPARATUS FOR DETECTING INTRUDER &

The present invention relates to an intrusion detection method and apparatus, and more particularly, to an intrusion detection method and apparatus using signal processing of raw data without using an FFT (Fast Fourier Transform).

In existing radar sensors, frequency characteristics are extracted from raw data converted to digital signals from a radio frequency (RF) stage through FFT (Fast Fourier Transform) to effectively distinguish between noise and valid signals, and SNR (Signal to Noise Ratio) and obtains the valid signal. The obtained valid signal is compared with an appropriate threshold value to determine whether an alarm is present or to extract desired data.

At this time, various noise such as hardware noise due to characteristic of RF (radio frequency) and noise due to installation environment exist, and radar sensors utilize various software methods to remove such noise, and a differential filter do.

That is, in most radar sensors, it is impossible to judge an effective signal by simply comparing raw data with an appropriate threshold value due to various noises. Therefore, the frequency characteristic is extracted from the raw data by FFT and signal processing is performed.

Such an FFT is complex and requires a large amount of computation. In the case of an embedded system, a high performance MCU (Micro Controller Unit) must be used to implement an FFT. If there is no separate FFT hardware DSP block in the MCU, It is virtually impossible to perform the FFT and subsequent signal processing.

Also, in the case of the widely used SNR for distinguishing the valid signal in the post-FFT signal processing, it is difficult to distinguish the signal to be detected according to the variation of the environmental noise, and the probability of false detection increases. For example, SNR at a location where there is little environmental noise is very useful for extracting the valid signal, but it is sometimes difficult or impossible to extract a signal with a small size where the environmental noise is large. The differential filter removes environmental noise and unwanted signals. It is complicated and requires a large amount of computation in a variety of situations, such as when the environmental change is severe, that is, when the influence of rain, wind, or temperature or humidity changes rapidly. Signal processing should be performed.

An object of the present invention is to provide an intrusion detection method and apparatus capable of acquiring an effective signal corresponding to an intrusion through signal processing of raw data, which is resistant to environmental noise and does not use expensive hardware.

According to one embodiment of the present invention, a method of detecting an intrusion in a radar sensor is provided. The intrusion detection method includes the steps of: collecting a digital I signal and a Q signal from a received RF signal; calculating an absolute value of a digital I signal and a Q signal, respectively; convolving an absolute value of the I signal and the Q signal; Calculating an average from the convolved signal for a set time, applying an exponential weighted filter function to the average to set a first threshold for valid signal detection, and comparing the average to the first threshold And detecting a valid signal for intrusion detection through the second interface.

The step of setting the first threshold value may include generating a reference value by applying an exponential weighting filter function to the average and setting the first threshold value to a value proportional to the reference value .

The intrusion detection method may further include determining whether to intrusion via comparison between an average of the signals detected as the valid signal and a second threshold value set for the intrusion detection.

The collecting may include down-converting the RF signal to an intermediate frequency signal, and converting the intermediate frequency signal from an analog signal to a digital signal.

Generating the reference value includes generating a reference value at the current time point by applying exponential first and second weights to the reference value generated immediately before and the average calculated at the current time point, , The sum of the first and second weights is 1, and can be changed according to environmental factors.

According to another embodiment of the present invention, there is provided an intrusion detection device for detecting an intrusion in a radar sensor. The intrusion sensing apparatus includes a receiver for acquiring a digital I signal and a Q signal from a received RF signal, an absolute value calculator for calculating an absolute value of the I signal and the Q signal, an absolute value calculator for calculating an absolute value of the I signal and the Q signal, An exponential weighting filter for generating a reference value by applying an exponential weighting filter function to the average, and an exponential weighting filter for extracting a valid signal from the reference value, And a signal detector for detecting a valid signal for intrusion detection by comparing the average and the first threshold value.

The signal detector may set the first threshold value to a value proportional to the reference value.

The intrusion detection apparatus may further include an intrusion determination unit for determining whether to intrusion through comparison between an average of the signals detected by the valid signal and a second threshold value set for detecting the intrusion.

Wherein the exponential weighting filter generates a reference value at the current point by applying exponential first and second weights to the reference value generated immediately before and the average calculated at the current point, The sum of the weights is 1, and can be changed according to environmental factors.

According to the embodiment of the present invention, since the FFT is not used, it can be immediately applied to existing radar sensors without expensive hardware configuration, and the probability of false detection and non-detection can be reduced by sufficient amplification of signals and minimization of environmental noise.

1 is a diagram schematically showing the configuration of an intrusion detection apparatus according to an embodiment of the present invention.
2 is a diagram showing an example of a raw data signal which is an output signal of the receiving unit shown in FIG.
3 is a flowchart illustrating an intrusion detection method according to an embodiment of the present invention.
4 is a diagram illustrating an example of a raw data signal converted into a digital signal and a signal passed through an average filter according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and similar parts are denoted by like reference characters throughout the specification.

Throughout the specification and claims, when a section is referred to as "including " an element, it is understood that it does not exclude other elements, but may include other elements, unless specifically stated otherwise.

Now, an intrusion detection method and apparatus according to an embodiment of the present invention will be described in detail with reference to the drawings.

1 is a diagram schematically showing the configuration of an intrusion detection apparatus according to an embodiment of the present invention.

1, the intrusion detection apparatus 100 includes a receiving unit 110, an absolute value calculating unit 120, a convolution calculating unit 130, an averaging filter 140, an exponential weighting filter 150, 160 and an intrusion determination unit 170. [ The intrusion detection device 100 may be a radar sensor itself or a radar sensor.

The radar sensor transmits a signal, receives a signal reflected from an object in the sensing area, and processes the received signal to detect intrusion in the sensing area.

Specifically, the receiving unit 110 converts the received RF signal into an in-phase signal and a quadrature signal of a digital signal after performing an intermediate frequency down-conversion. And outputs the I signal and the Q signal converted into the digital signal to the absolute value calculation unit 120.

The absolute value calculator 120 calculates absolute values (ABS.I and ABS.Q) of the I signal and the Q signal, respectively, and outputs the absolute values (ABS.I, ABS.Q) of the I signal and the Q signal to convolution And outputs it to the calculation unit 130.

The convolution operation unit 130 convolutes and outputs the absolute value (ABS.I) of the I signal and the absolute value (ABS.Q) of the Q signal for the set time. The convolution can be calculated as shown in Equation (1).

That is, as the magnitude of the signal received by the receiving unit 110 increases, the absolute values (ABS.I, ABS.Q) of the I signal and the Q signal become larger and the absolute value (ABS.I) of the I signal and the absolute value (Effective signal due to intrusion or the like) is amplified by taking the convolution of the signal (ABS.Q), it is possible to easily distinguish the signal change due to intrusion or the like.

The average filter 140 calculates an average value from the convolved signal Conv for a predetermined time and outputs it to the exponential weighting filter 150 and the signal determining unit 160. This average filter 140 removes white noise due to various environmental factors from the convolved signal Conv.

Due to the characteristics of the radar sensor, the size of the radio wave always changes according to the installation environment even when there is no valid signal, and it is difficult to extract the change of the effective signal due to the change of the radio wave size according to the change of the environment. Therefore, in the embodiment of the present invention, an exponential weighting filter 150 is used. An exponential weighting filter 150 is used to adapt the background signal to be compared with the valid signal according to the environment.

)을 생성한 후, 기준 값을 신호 결정부(160)로 출력한다. 지수 가중 필터 함수는 수학식 2와 같이 표현될 수 있다. The exponential weighting filter 150 applies an exponential weighting filter function to the signal output from the averaging filter 140 to extract a reference value

And outputs the reference value to the signal determination unit 160. [ The exponential weighting filter function can be expressed as Equation (2).

는 현재 시점(k)에서의 기준 값으로서, 가중치를 적용한 추정값이다.

는 직전 시점(k-1)에서의 추정값이며,

는 현재 시점(k)에서의 측정값으로, 평균 필터(150)로부터 출력된 평균 신호이다. α는 가중치로서, 지수함수적인 상수이다. α는 환경 요인에 따라 변경될 수 있다. here,

Is a reference value at the current point (k), and is an estimated value to which a weight is applied.

Is an estimated value at the immediately preceding time point (k-1)

Is an average signal output from the averaging filter 150 as a measured value at the current point (k). α is a weight and is an exponential constant. α can be changed according to environmental factors.

At this time, if the value of [alpha] is decreased, the original data signal becomes insensitive to the environment change, and the whole system becomes sensitive. If the value of [alpha] is increased, the original data signal becomes sensitive to environmental change and the whole system becomes insensitive. ? is set to be larger than the unit number in the average filter.

As described above, in the embodiment of the present invention, by applying the exponential weighting filter 150 using the recursive equation, the primitive data signal can be effectively adapted to the change of the environment.

)와 지수 가중 필터(150)로부터 출력된 기준 값(

)을 비교하여, 유효 신호를 판단한다. 구체적으로, 신호 결정부(160)는 지수 가중 필터(150)로부터 출력된 기준 값(

)을 토대로 임계값을 설정하고, 평균 필터(140)로부터 출력된 신호(

)와 설정된 임계값을 비교하여 유효 신호를 판단한다. 임계값은 평균 필터(140)로부터 출력된 신호(

)에 비례한 값으로 결정될 수 있으며 이때 적용되는 비례 계수는 해당 시스템에 따라서 적절하게 지정될 수 있다. The signal determining unit 160 receives the signal (

) And an exponential weighting filter 150

) To judge the valid signal. Specifically, the signal determination unit 160 determines a reference value (a reference value) output from the exponential weighting filter 150

And outputs the signal output from the average filter 140 (

) And the set threshold value to determine the valid signal. The threshold value is a signal ("

), And the proportional coefficient applied at this time can be appropriately designated according to the system.

)를 침입 감지를 위해 설정된 임계값과 비교하여 침입 여부를 결정한다. The intrusion determination unit 170 determines whether or not the signal

) With the threshold value set for intrusion detection to determine whether or not to intrusion.

2 is a diagram showing an example of a raw data signal which is an output signal of the receiving unit shown in FIG.

As shown in FIG. 2, both the I signal and the Q signal converted into the digital signal have both positive and negative numbers. The larger the magnitude of the signal received by the receiver 110, the larger the absolute value of the I signal and the Q signal. In addition, the amount of the received radio wave varies according to the change of the object in the intrusion detection area. At this time, the signal changes abruptly and is separated from the impulsive noise signal.

3 is a flowchart illustrating an intrusion detection method according to an embodiment of the present invention.

Referring to FIG. 3, the radar sensor receives an RF signal (S310), performs intermediate frequency down-conversion, and then converts the analog signal into an I signal and a Q signal of a digital signal (S320).

The radar sensor calculates an absolute value of the I signal and the Q signal converted into the digital signal (S330), and then, the absolute value of the I signal and the Q signal is convoluted (S340).

The radar sensor calculates an average from the convolved signal for a predetermined time (S350), and applies an exponentially weighted filter function to the calculated average to generate a reference value for threshold setting (S360).

The radar sensor sets a threshold value to a value proportional to the reference value (S370), compares the average calculated from the convoluted signal with a threshold value, and detects whether the corresponding signal corresponds to an effective signal due to intrusion or the like (S380).

The radar sensor determines whether the intruder is intruding by comparing the average of the signals detected as the valid signal with the handover value set for intrusion detection (S390).

4 is a diagram showing an example of a raw data signal converted into a digital signal and an effective signal detected according to an embodiment of the present invention.

)를 나타내고, (B)는 본 발명의 실시 예에 따른 지수 가중 필터(150)로부터 출력되는 기준 값에 비례하여 설정된 임계값을 나타낸다. The left graph of FIG. 4 shows a raw data signal converted into a digital signal, and in the right graph of FIG. 5, (A) shows a signal output from the average filter 140 according to the embodiment of the present invention

(B) represents a threshold value set in proportion to the reference value output from the exponential weighting filter 150 according to the embodiment of the present invention.

Referring to FIG. 4, in the original data signal converted into the digital signal, there exist signals having a size similar to that of the effective signal even in an area other than the effective signal area. Therefore, it is difficult to determine intrusion detection by comparing a raw data signal converted into a digital signal with a threshold value set for intrusion detection.

)와 설정된 임계값을 비교함으로써, 유효 신호를 결정할 수 있다. 이렇게 하면, 유효 신호 영역에서만 유효 신호가 검출되고, 유효 신호 영역 이외의 영역에서는 유효 신호가 검출되지 않는다. However, as shown in the right graph of FIG. 4, the threshold value is set in proportion to the reference value output from the exponential weighting filter 150, and the signal output from the average filter 140

) And the set threshold value, the valid signal can be determined. In this way, the valid signal is detected only in the valid signal area, and no valid signal is detected in the area other than the valid signal area.

As described above, according to the embodiment of the present invention, the signal amplification using the convolution, the noise removal using the averaging filter 140, and the exponential weighting filter 150 The validity signal can be effectively detected only for the valid signal region.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, It belongs to the scope of right.

Claims (9)

A method of detecting an intrusion from a radar sensor,
Collecting a digital I signal and a Q signal from the received RF signal,
Calculating the absolute values of the digital I signal and the Q signal, respectively,
Convolving absolute values of the I signal and the Q signal;
Calculating an average from the convolved signal for a set time,
Applying an exponential weighting filter function to the average to set a first threshold for effective signal detection, and
Detecting an effective signal for intrusion detection by comparing the average and the first threshold value
Wherein the intrusion detection method comprises the steps of: The method of claim 1,
The step of setting the first threshold value
Applying an exponential weighting filter function to the average to generate a reference value, and
And setting the first threshold value to a value proportional to the reference value. The method of claim 1,
Determining whether to intrusion through comparison of an average of signals detected with the valid signal and a second threshold value set for detecting the intrusion;
Further comprising: The method of claim 1,
The collecting step
Down converting the RF signal into an intermediate frequency signal, and
And converting the intermediate frequency signal from an analog signal to a digital signal. 3. The method of claim 2,
Generating the reference value includes generating a reference value at the current time point by applying exponential first and second weights to the reference value generated immediately before and the average calculated at the current time point, ,
Wherein the sum of the first and second weights is 1 and is changed according to environmental factors. An intrusion detection device for detecting an intrusion from a radar sensor,
A receiver for collecting the digital I signal and the Q signal from the received RF signal,
An absolute value calculator for calculating an absolute value of the I signal and the Q signal,
A convolution operation unit for convoluting absolute values of the I signal and the Q signal,
An average filter that calculates an average from the convolved signal for a predetermined time,
An exponential weighting filter for generating a reference value by applying an exponential weighting filter function to the average, and
A signal detector for detecting a valid signal for intrusion detection by setting a first threshold value for detecting an effective signal from the reference value and comparing the average value with the first threshold value,
And an intrusion detection device. The method of claim 6,
Wherein the signal detector sets the first threshold value to a value proportional to the reference value. The method of claim 6,
And an intrusion determination unit for determining whether to intrusion through comparison between an average of the signals detected as the valid signal and a second threshold value set for detecting the intrusion,
The intrusion detection apparatus further comprising: The method of claim 6,
Wherein the exponential weighting filter generates a reference value at the current point by applying exponential first and second weights to the reference value generated immediately before and the average calculated at the current point,
Wherein the sum of the first and second weights is 1, and is changed according to environmental factors.

Images